Severe endothelial rejection was treated with topical and systemic steroids in all cases and the rate of reversibility was 63.8%. The rate of reversibility was reported to be 51–63.3% in previous studies [7, 8]. Wagoner et al. investigated 152 cases of severe endothelial rejection following PKP treated with topical steroids with or without systemic steroids and reported that graft survival was 42.6% at 1 year and 36.1% at 3 years . Although our series was relatively small and some differences existed among other factors, the effectiveness of our treatment regimen in severe cases was comparable with that of earlier report. Our results suggest additional systemic steroids provide similar effective outcomes even in severe cases with severe edema. However, it is difficult to compare the status of patients before treatment with previous reports rigorously, and further study is needed. No serious systemic side effects were observed because we excluded patients in poor general health. Careful observation to detect elevated IOP and infection, especially herpes simplex or fungal keratitis, is necessary when treating patients with topical or systemic steroids.
In this study, a longer interval between corneal graft rejection and treatment with systemic steroids was associated with an increased risk of corneal decompensation after graft rejection. Risk factors for irreversibility after graft rejection reported in previous studies included donor age, patient age, diagnosis of BK, history of rejection or graft failure episodes [7, 9]. Early treatment was reported to be associated with a better outcome  and our results support this finding. Factors affecting corneal decompensation after the recovery of corneal transparency were also investigated. Corneal decompensation occurs in one-third of the cases within approximately 6 months. A comparison of these cases with those in which corneal transparency was maintained revealed that regraft as a diagnosis before previous PKP was more frequent in the former. Notably, this factor was not associated with graft reversibility of transparency, and factor affecting graft reversibility of transparency was not associated with the maintenance of graft transparency. In regraft cases, more careful observation is needed after corneal transparency has been restored.
The endothelial cell density decreased significantly after corneal graft rejection. Musch et al. reported that the ECD decreased by 11.8% , while we observed a reduction in ECD in 18 cases in which ECD was determined before and after corneal graft rejection for a rate of 50.4%, which was much higher than that in Musch et al.. Furthermore, cases in which ECD could be calculated, indicating the absence of severe edema, were believed to be mild cases compared with cases in which ECD could not be calculated in our study. Therefore, the endothelial cell loss may have been underestimated. These indicate the greater incidence of severe cases of rejection in our study compared with previously reported series . Our results suggest the importance of preventing rejection, as well as close monitoring and appropriate and aggressive management of rejection when it occurs.
The interval between PKP and corneal graft rejection was 31.5 ± 36.7 months, which was longer than that in previous studies, including 15.4 ± 20.9, 10.4 ± 9.3 and 15.3 ± 14.4 months reported respectively by Epstein et al., Naacke et al. and Sangwan et al.. One reason for this discrepancy between our results and those of these earlier studies may be that topical steroid treatment after PKP tended to be continued longer in our patients. In fact, we recently reported the efficacy of prolonged use of topical steroids for the prevention of rejection after PKP . If no side effects are observed, such as elevated IOP, cataracts or infection, the long-term use of topical steroids is recommended.